ANNIE DUKE: What 'resulting' is using the outcome quality as a perfect signal for deriving decision quality. So, let me give you an example. It's 2015; people will remember the Super Bowl the Seahawks against the Patriots and the Seahawks are on the one-yard line, there's 26 seconds left, it's the second down and they have one timeout. And I think that people will remember that - and they're down by four by the way - people will remember that famously Pete Carroll called a pass play. Russell Wilson passed the ball. It was intercepted. And the next day the headlines were let's just say pretty bad for Pete Carroll. Worst play in Super Bowl history; Pete Carroll I think with some of them called him an idiot, but headline after headline after headline after headline was about how completely ridiculous this call was.
Now, there were a couple of outlying voices, one of the main ones was a guy named Benjamin Morris over at FiveThirtyEight and he went through some of the analytics on the play and actually had very, very good arguments for why that was at worst reasonably thought through and possibly quite a brilliant play if you were just thinking statistically. It's easy to see what's happening here because all you have to do is the thought experiment. And he thought experiment is this: Pete Carroll calls to pass; Russell Wilson throws it is; and it's caught in the end zone for a touchdown. And just take a minute and think about what those headlines would have been the next day. Instead of worst play in Super Bowl history they would've been Carol outsmarts Belichick, it would've been about his creativity this is the kind of thinking that got him to the Super Bowl in the first place. This is why he's the best coach in the NFL and deserves that ring. Now, obviously whether the ball is caught or dropped does not actually change whether the decision was a good one, but we act like it does and that's what 'resulting' is.